A REVIEW OF RESEARCH PROGRESS OF PREDICTION MODELS FOR DISINFECTION BY-PRODUCTS:EMPIRICAL MODELS
Disinfection by-products(DBPs)are the reaction products during the disinfection process of drinking water,which are a serious threat to human health.Therefore,it is crucial to establish relevant models to predict their concentrations and achieve accurate control.This paper reviews the research progress of empirical models for DBPs prediction,briefly reviews the current common disinfection means,types of DBPs,and the corresponding relevant norms and standards,and explores the principles of DBP models based on regression and machine learning,respectively.The prediction effects of models constructed by taking these two approaches are summarized and evaluated.Among them,the principles of machine learning algorithms for three DBPs prediction models,namely,random forest algorithm,support vector machine,and artificial neural network,are focused on and analyzed.This paper puts forward the problems of the current DBPs disinfection by-products prediction model.It looks forward to its future development direction,aiming to promote the building of the prediction model with higher accuracy and applicability.
NETL
NSTL
万方数据
